Knowledge Amplification through Randomization for Scheduling Systems

Case-Based Reasoning (CBR) is an analogy-based method allowing for resolving new problems by exploiting previously accumulated knowledge and experiences. Randomization is a novel approach for knowledge generation and compactification. Randomization improves CBR when integrated with it. The work presented in this paper pertains to knowledge amplification based on randomization. New knowledge is deduced from hidden knowledge by subsumption. The approach is applied to a scheduling system, thus highlighting its strength in enhancing case-based reasoning by inferring pertinent new and valid knowledge. Experimental results show the efficiency of the approach.